Study area description
Anopheles larval sampling was made from the potential breeding habitats in the Arba Minch area, located in the southern Ethiopian Rift Valley system. The climate is hot and humid which is favourable for mosquito breeding and malaria transmission. The average maximum temperature is 30°C, while the average minimum is 17°C. The rainfall distribution is bimodal and mostly occurring in March to May and September to November. The annual rainfall average is around 900 mm. The major cash crops cultivated in the area include bananas, maize and mango using both rain and irrigation schemes. In the rainy seasons, the water pools in ditches along the roads, farmlands, tire track pooled water, and marshy areas are the potential breeding sites for Anopheles mosquitoes.
Malaria is a common health problem in the area13,11 and is more prevalent after the end of the rainy season due to the potential breeding sites11. Plasmodium falciparum and P. vivax are the two common species14,15.
Study design
This experimental study was conducted to assess the effect of pollen of some selected plants on the larval survival, pupation rate, adult emergence, vector longevity and infectivity to malaria parasites. Newly emerged larvae were provided with equal quantity of pollen from T. latifolia, Z. mays, P. juliflora and standard larval food Tetramin fish food (Cichlid Sticks; Tetra, Maidenhead Aquatics, Leicester, UK). The emerged adults were supplied with10% sugar solution until ready for the infection experiments16. Then, 2–3 days old female Anopheles were starved overnight and provided with P. vivax and P. falciparum parasite infected blood though membrane feeding (Fig. 1).
Study plants
These plants (Fig. 2) were selected based on the scientific evidences documented in relation to the malaria mosquitoes either against or in favour of larval and adult stages5,7. Typha latifolia (common cattail) is a perennial herbaceous plant in the genus Typha and among the most common of all aquatic plants. It is an "obligate wetland" species found close to water bodies. This plant is among the wind-pollinated plants that produces large quantities of pollen. Prosopis juliflora is a shrub or small tree in the family Fabaceae. It is an evergreen and fast-growing invasive species in arid and semi-arid areas. The flowers of P. juliflora are self-incompatible, entomophilous (insect pollinating), depending on insects for seed dispersal. It produces many seeds and is tolerant of a wide range of climatic regimes and soil types which have contributed to making P. juliflora one of the worst invasive alien plants. The leafy stalk of Z. mays (maize) produces pollen inflorescences and separate ovuliferous inflorescences called ears that yield seeds, which are fruits and its pollen, are produced entirely in the staminate inflorescence that is wind pollinated. The male flower matures earlier than the female flower. The plant specimens were collected from their typical habitat by field press and brought to the Arba Minch University Botany laboratory for further confirmation by expertise.
Pollen collection
Pollen of T. latifolia was collected in the shore of the Lake Abaya, P. juliflora from the shrub dominated area of Lante (local place in Arba Minch area) and the pollen of Z. mays L. (BH660 cultivar) from the farmland of Lemat during the maize flowering season. The pollens were collected by covering the mature flower with a paper bag and shaking. The collected pollen was transferred to vials and sterilized within 24 hours of collection using UV light (10850 Harry Hines Blvd Suite 155, Dallas, TX 75220) overnight for 12 hours and store on silica gel until used for feeding experiments.
Pollen nutrient analysis and measuring grain size
The C and N concentration of pollen diets were determined by Walkley-Black and Kjeldahl methods in Arba Minch University analytical chemistry laboratory. Approximately 5 mg of sample of pollen from each plant was taken and processed based on the standard protocol. Thereafter, the C: N ratio of the pollen was determined by %C/%N × 6.2517. The pollen grain sizes were measured according to Roulston et al17. Pollen grains were mounted in paraffin oil and measured at 400× with an ocular micrometer. Polar axis measurements were taken for all pollen grains. Volumes were calculated using volume equations for spheres (4/3πp3). Ten randomly selected grains per slide were measured, and these values were averaged for calculating volumes.
Mosquito larvae collection and rearing
Anopheles larvae were collected from the potential breeding habitats in the Arba Minch area and reared in Arba Minch University Medical Entomology and Vector Control laboratory. The rearing room was maintained at temperature between 25–27°C and relative humidity between 72–80%. The larvae rearing trays were inspected twice a day for the presence of pupae, and if present they were placed in adult cages. The newly emerging adults were provided with 10% sterilized sugar solution. Then, they were allowed feed on live rabbit at 3- to 4-days interval and serving as ‘mother’ cages. The egg dish was placed in the cage one day after blood feeding and eggs were transferred to larvae tray for hatching. These newly hatched first instar larvae were used for the experimental feeding of pollen diets. The female adult mosquitoes in the mother cage were morphologically identified for An. gambiae complex and a sub-sample (150) was confirmed by polymerase chain reaction (PCR) for An. gambiae complex. The entire tested An. gambiae complex was identified as An. arabiensis (hereafter the target mosquito species will be An. arabiensis).
Larval development and survival analyses
After 12 hours of hatching, 200 unfed first instar larvae were counted manually and randomly apportioned to larval trays of 10×25 cm and 8 cm high filled with one litter of distilled water. Effective dose of pollen diet and TetraMin fish food was determined by optimizing 0.15, 0.3 and 0.6mg on relative adulthood survival. Then, 0.3mg was established as the effective dose of each pollen diet and Tetramin fish food.
Six replicates each with 25 larvae for each of the plant pollen diet and two as control group were used. The standard TetraMin fish food was used as control. The experiment was repeated three times for each pollen diet and Tetramin fish. A total of 450 larvae were treated for each diet and 150 for each experimental term. For the first and the second instar larvae, the diet was offered only once a day, for the third and fourth, it was twice a day based on procedure of Kivuyo et al18. In each replicate, the larval development and survival was recorded twice a day (01:00 and18:00) and dead individuals were counted and removed. Continuous inspection of the number of larvae surviving, pupating, and emerging adults were recorded.
Survival rate of each day (sj) =
Survival rate total (S) = s1 + s2……+sj
Where, s1 the no. of larvae at the begning of day 1
s2 the no. of larvae at the begning of day 2
Measuring Mosquito Size/wing Length
The same batches and same ages of 30 female adult An. arabiensis were collected from each pollen diet and TetraMin fish food. They were killed by freezing and, then one wing of each mosquito glued onto a slide. Its length from the distal end of the alula to the tip, excluding the fringe scales, was measured with stage graticule micrometer (ruler etched in side) to examine the effect of larval diet on wing length of adult An. arabiensis.
Patient screening for gametocyte stage
Microscopically positive patients with P. vivax and P. falciparum gametocytes were enrolled. The gametocyte detection was done by standard finger prick and Giemsa stained blood smears from thick blood smear under oil immersion microscopy by experienced laboratory technicians. The gametocyte positive patients were asked to provide approximately 3ml of blood for membrane feeding prior to antimalarial treatment. Blood gametocyte density was estimated microscopically by converting the total number of gametocytes in 1000 leukocytes to 8000 leukocytes/µl. Screening of gametocyte carriers was synchronized into mosquito rearing processes and thus always performed on the same day with the experimental infection. Patients with mixed infection were excluded.
Membrane blood feeding
The same batch of 2 to 3-day-old female An. arabiensis were starved overnight and kept for experimental infection through membrane feeding. A total of 1280 female An. arabiensis were allowed to feed on P. vivax positive blood in four membrane feeding experiments, and another 1280 were exposed to P. falciparum positive blood in four membrane feeding experiments. In each feeding experiment, 320 An. arabiensis were exposed in 8 paper cups each with 40 mosquitoes.
The circulation water bath was used to allow water circulation through the feeder and its temperature was adjusted on 37°C to keep the blood warm and the parasites alive. A Plasmodium-positive blood meal was carefully pipetted from the same blood source into the neck of the feeder and allowed to feed simultaneously for 30 minutes under dark conditions. After carefully looking at the mosquito for abdominal condition, unfed mosquitoes were discarded. The fed An. arabiensis were kept in the paper cups and provide with 10% sugar solution under normal insectary condition (27°C ± 2 and 70 ± 10% relative humidity). Anopheles arabiensis were kept under sugar for eight days for P. vivax and twelve days for P. falciparum oocyst detection. Infected mosquito containing cups were kept in a small cage for double protection. The membrane was removed carefully and soaked in 10% bleach to decontaminate and rinse with water.
Dissection to oocyst stage and sporozoite detection
Randomly 40 mosquitoes were taken from each pollen type replicate cages on day 7 post-infection for P. vivax and on day 12 post infection for P. falciparum. A total of 120 female mosquitoes were dissected for P. vivax and P. falciparum. For each of the dissection sessions, the infection rate (IR) (infected mosquitoes/total mosquitoes dissected in each feeding (%) was recorded, and intensity of infection (oocyst load/total infected mosquitoes dissected in each feeding) were determined. Non-infected mosquitoes were omitted from the calculation.
The infective stage of the Plasmodium parasite was detected by circumsporozoite protein (CSP)-ELISA using the kits specific for P. vivax-210, P. vivax-247 and P. falciparum following standard instruction19. A total of 104 and 35 An. arabiensis were tested for CSPs P. vivax and P. falciparum, respectively.
Data analysis
The effect of diets on larval survival was performed, and to conduct survival analysis data were converted to count-time data. Kaplan-Meier curves were used to compare diet-specific survival pattern for each group. Comparison of survival rate in each larval diet to that of TetraMin Fish food was performed using log rank test. ANOVA was used for pupation, emergence rate and wing length to assess the influence of main variables (larval diet). Wald’s test was used to determine significant differences between diets based on χ2 and P-values.
The proportion of female adult mosquitoes that survived after infected blood meal determined by counting the number of mosquitoes on their oocyst and sporozoite stage across all larval diets. The median number of oocyst per mosquito mid-gut was compared using a Kruskal-Wallis test, followed by a Dunn's post-hoc test. The prevalence of sporozoite infection, or percent of mosquitoes reaching to infective stage, was compared with a Fisher's exact test.
All statistical analysis performed using GraphPad Prism 5 (Version 5.01). Significance was considered at 5% level of confidence. Single factor experimental design was employed to assess the impact of pollen of some selected plants on the larval survival, pupation, adult emergence, and vector infectivity to malaria parasites. Survival rate of each treatment case was calculated as survival rate of each day (Araujo et al., 2012).
Ethical Consideration
This study was reviewed and approved by the Institutional Review Board of Arba Minch University (No: CMHS/12036564/111). After briefing the objective of the study, consent was obtained from the study participants. Participants in the study who tested positive for Plasmodium received free treatment in accordance with the national malaria treatment guidelines. The CSP-ELISA test was performed in accordance with standard protocol19. All methods were performed in accordance with the standard protocols and procedures18,19. Other plant pollens, aside from maize pollen, were collected with the local administration approval in open spaces like the lakeshore and by the side of the road. Through permission of farm owners, the maize pollen was collected on farmland. Despite the fact that only the pollen grains were gathered for the experiment, plant specimens were also gathered in order to be identified by senior botanists before collection. The voucher specimens of the plants (P. juliflora and T. latifolia) have been deposited in a publicly available botanical herbarium at Arba Minch University.